Unverified50% confidenceBenchmarkExact time
On CIFAR-100 (ResNet-18), SALU achieves 83.42% Top-1 at BS=512 and still retains 76.23% Top-1 at BS=1, whereas standard BatchNorm collapses at BS=1.
1
Sources
50%
Confidence
Long-term
Relevance
11/9/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen trained from scratch on ImageNet-1K, SALU reaches 80.16% Top-1 with ResNet-50 and 80.28% Top-1 with ViT-B/16, using only BS=128.72% similarUnverifiedThe Top3 configuration achieved an accuracy of 97.148%, a weighted F1 of 97.055%, and an MCC of 0.9675, comprehensively outperforming the Full21 configuration using all 21 metrics (accuracy 95.999%, weighted F1 95.521%, MCC 0.9549).63% similarVerifiedLog4Shell (CVE-2021-44228) received a perfect CVSS score of 10.0.60% similarUnverifiedOn a subset containing only 249 home/catalog images, Pasma's Top-1 accuracy was 83.53% versus 87.15% for SigLIP-Base, reversing the ranking.59% similarUnverifiedFor annotation consistency, Cohen's kappa was 0.781 (range 0.753-0.807) and Fleiss' kappa was 0.77.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1151068API
curl https://kongchang.com/api/v1/knowledge/claims/1151068MCP
get_claim(id=1151068)